docx-processor

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from Microsoft Word files, creating a potential surface for indirect prompt injection.
  • Ingestion points: Document content is read and extracted in scripts/read_docx.py and scripts/convert_docx.py.
  • Boundary markers: None identified; extracted text and table data are returned directly to the agent context without delimiters or instructions to ignore embedded commands.
  • Capability inventory: The skill can read local files and write converted output to the /tmp/openskills-converted directory. It does not have network access or arbitrary command execution capabilities.
  • Sanitization: None; document content is extracted and processed as raw text.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 02:59 PM
Security Audit — agent-trust-hub — docx-processor